Melanoma is an aggressive form of skin cancer whose overall incidence and mortality rates are increasing worldwide. There is significant unresolved complexity in the clinical and histopathological presentation, relationship to UV light, and epidemiology of melanoma. Attempts to stratify tumors into "histogenetic" types based on the microscopic pattern of the disease have not been broadly accepted in the clinic because of significant overlap between the criteria that are proposed and limited relationship between the classification and clinical course of the disease. Over the past decade our studies of primary melanomas have resulted in the demonstration of genetically distinct subtypes of melanoma that are related to patient germline genotypes, anatomic site of presentation, and to solar exposure. This work has resolved the issue of whether there are meaningful distinctions to be made among melanomas, and moved the field to address the discovery of those differences. Additional progress in this area is critical since current therapies have proven ineffective, perhaps in part because responses in subsets of patients are not recognized. The goal of this proposal is to establish a framework of genetic and phenotypic markers that define disease categories with common mechanistic alterations and to provide markers that can be used for classification, diagnosis, and to identify groups of patients that are likely to respond to emerging therapies targeted against genetic alterations. To achieve this goal we will assemble a phenotypically diverse cohort of 350 primary melanomas by sampling from categories of the existing WHO classification. This discovery and training cohort will used to identify new genetic markers, refine phenotypic assessments and establish classification algorithms based on the genetic, clinical and histopathological characteristics. The provisional classification algorithms will be validated for their ability to produce genetically homogeneous subgroups using an independent cohort of primary tumors, and the association of these subgroups with disease outcome will be determined using a third cohort. Finally we will define surrogate markers using only commonly collected clinical information, and apply them to the 50,000 cases in the AJCC tumor registry to determine if they provide improved prognostic power compared to current practice. PUBLIC HEALTH RELEVANCE: Melanoma is an aggressive form of skin cancer comprised of still incompletely defined subtypes. The goal of this project is the discovery of markers that improve classification, diagnosis, and stratification for therapy.